@article{ref1,
title="Falling analysis and examination of different novel strategies for preserving the postural stability of a user wearing ASR-EXO during stair climbing",
journal="Journal of intelligent and robotic systems",
year="2022",
author="Borooghani, Danial and Hadi, Alireza and Alipour, Khalil and Tarvirdizadeh, Bahram",
volume="105",
number="1",
pages="e5-e5",
abstract="Falling of aged or disabled people during stair climbing, when using assisting devices, can be dangerous. Falling analysis and postural stability algorithms based on it can abstain the subjects from inflicting those injuries. In this research, a subject using ASR-EXO, a newly developed lower-limb exoskeleton, has been modeled and co-simulated with MSC ADAMS and Matlab. To posit novel postural stability strategies, Centroidal Moment Pivot (CMP) is utilized for a person wearing ASR-EXO. Further, a new strategy of balance named Trunk's Angular Jerk (TAJ) is introduced. The strategies are implemented in MSC ADAMS Matlab co-simulation using a PD controller. When the subject ascends stairs, four phenomena of leaning, abnormal heel contact, toe abrasion, and toe-stair contact happen that can be used as four signs indicating the direction of falling. They can imply falling forward or backward. The four phenomena would leave their impact on the amplitude of the peak of Trunk's Translational Acceleration (TTA), Trunk's Angular Acceleration (TAA), and time derivative of TTA. Based on a comparison between these peaks during balanced and unbalanced stair climbing, the Trunk's Translational and Angular Acceleration (TTAA) algorithm is introduced. It defines safe boundaries on the magnitude of these peaks, based on simulation and experimental data and consequently, relates the four phenomena to the falling directions. Moreover, the TTAA calculates the right time for Postural Stability Torques (PSTs) to be exerted by ASR-EXO and keeps the mentioned parameters in their safe domains. The TTAA algorithm can guarantee the Postural Stability of the subject with only a torque of 6 N.m, instead of 30 N.m and 200 N.m of CMP and TAJ algorithms respectively.<p /> <p>Language: en</p>",
language="en",
issn="0921-0296",
doi="10.1007/s10846-022-01629-w",
url="http://dx.doi.org/10.1007/s10846-022-01629-w"
}